Hydrogenation of 3-nitroacetophenone over rhodium/silica catalysts: Effect of metal dispersion and catalyst support

Article abstract of DOI:10.1016/j.apcata.2013.05.002

The effect of physical properties of rhodium/silica catalysts on the hydrogenation of 3-nitroacetophenone was investigated using a series of catalysts with different particle sizes, pore sizes and dispersions. The hydrogenation reaction was carried out in the liquid phase at different temperatures (313-343 K) and hydrogen pressures (2-5 barg). The hydrogenation of 3-nitroacetophenone was found to be a consecutive reaction with 3-aminoacetophenone, 1-(3-aminophenyl) ethanol, and 1-(3-aminocyclohexyl) ethanol the main products; traces of 3-aminoaniline were also detected. The reaction was zero order with respect to the concentration of 3-nitroacetophenone and half order with respect to hydrogen. The rate of formation and disappearance of intermediates increased with increasing temperature and hydrogen pressure. The rate of ring hydrogenation to form 1-(3-aminocyclohexyl) ethanol shows a marked increase at high temperatures. The catalytic activity varies with both the pore size of silica support and the rhodium dispersion with no significant influence of catalyst particle size. The turnover frequency increases with decreasing metal dispersion (or with increasing average metal crystallite size) revealing an antipathetic particle size effect. This behaviour can be related to changes in the electronic and structural make up of small metal crystallites but suggests that the reaction takes place on the plane face surface of the FCC close-pack structure of rhodium in which the number of face surface atoms increases at the expense of the edge and corner sites as the metal crystallite size increases.

Full text of DOI:10.1016/j.apcata.2013.05.002

Applied Catalysis A: General 462–463 (2013) 121–128
Contents lists available at SciVerse ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Hydrogenation of 3-nitroacetophenone over rhodium/silica catalysts:
Effect of metal dispersion and catalyst support
Majid I. Abdul-Wahab1, S. David Jackson∗
Centre for Catalysis Research, WestCHEM, School of Chemistry, University of Glasgow, Glasgow G12 8QQ, Scotland, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
The effect of physical properties of rhodium/silica catalysts on the hydrogenation of 3-nitroacetophenone
was investigated using a series of catalysts with different particle sizes, pore sizes and dispersions. The
hydrogenation reaction was carried out in the liquid phase at different temperatures (313–343 K) and
hydrogen pressures (2–5 barg). The hydrogenation of 3-nitroacetophenone was found to be a consecutive
reaction with 3-aminoacetophenone, 1-(3-aminophenyl) ethanol, and 1-(3-aminocyclohexyl) ethanol
the main products; traces of 3-aminoaniline were also detected. The reaction was zero order with respect
to the concentration of 3-nitroacetophenone and half order with respect to hydrogen. The rate of forma-
tion and disappearance of intermediates increased with increasing temperature and hydrogen pressure.
The rate of ring hydrogenation to form 1-(3-aminocyclohexyl) ethanol shows a marked increase at high
temperatures. The catalytic activity varies with both the pore size of silica support and the rhodium
dispersion with no significant influence of catalyst particle size. The turnover frequency increases with
decreasing metal dispersion (or with increasing average metal crystallite size) revealing an antipathetic
particle size effect. This behaviour can be related to changes in the electronic and structural make up of
small metal crystallites but suggests that the reaction takes place on the plane face surface of the FCC
close-pack structure of rhodium in which the number of face surface atoms increases at the expense of
the edge and corner sites as the metal crystallite size increases.
Received 12 March 2013
Received in revised form 29 April 2013
Accepted 3 May 2013
Available online xxx
Keywords:
Hydrogenation
3-Nitroacetophenone
Rhodium/silica
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
2,680,136 in 1954 [3], US patent 2,683,745 in 1954 [4], US patent
2,797,244 in 1957 [5], US patent 3,423,462 in 1969 [6], and US
The catalytic hydrogenation of 3-nitroacetophenone (3-NAP)
to its corresponding products is an important reaction in the
food and pharmaceutical industry. This hydrogenation process
has not been the subject of significant study and is not covered
well in the literature. The expected products mentioned in the
available literature are 3-aminoacetophenone (3-AAP) and 1-(3-
aminophenyl) ethanol (3-APE). 3-AAP is used as food flavouring
and in synthesising pharmaceutical intermediates such as adrianol,
and zaleplon [1]. 3-APE, is a useful intermediate in the synthe-
sis of dyestuffs: it may be condensed with haloanthraquinones
to yield arylaminoanthraquinones which, when devoid of water
solubilising groups such as sulfo and carboxy, find utility in the
field of polyester dyestuffs [2]. Early studies related to this reac-
tion are to be found in the US patent literature namely, US patent
patent 4,021,487 in 1977 [2]. Most of these patents focused on
the hydrogenation of 3-NAP to 3-AAP and sometimes to the alco-
hol, at relatively high pressures and temperatures using palladium
or Raney nickel as catalysts. Hawkins et al. [7] studied the hydro-
genation of 4-NAP to the corresponding amine, alcohol and ethyl
aniline products via a parallel reaction screening using different
palladium, platinum and rhodium catalysts. More recently, Jack-
son, et al. [1] has examined the hydrogenation of ortho, meta and
para-nitroacetophenone to the respective aminoacetophenone in
the liquid phase over a Pd/C catalyst. None of the previous work, to
the best of our knowledge, gave the full profile of this reaction.
In the present study, the hydrogenation of 3-NAP was investi-
gated over six rhodium/silica catalysts different in their physical
properties using a semi-batch reactor operated at different condi-
tions of pressure, temperature, and catalyst loading. In the present
hydrogenation process, 3-NAP was hydrogenated consecutively
to 3-aminoacetophenone (3-AAP), 1-(3-aminophenyl) ethanol (3-
APE), and 1-(3-cyclohexylamine) ethanol (3-ACHE), in addition to
small amounts of 3-ethylaniline. In a subsequent paper a full kinetic
analysis of the reaction sequence will be presented.
∗
Corresponding author. Tel.: +44 141 330 4443.
E-mail address: david.jackson@glasgow.ac.uk (S.D. Jackson).
Present address: Chemical Engineering Department, University of Baghdad,
1
Baghdad, Iraq.
0926-860X/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2013.05.002

122
M.I. Abdul-Wahab, S.D. Jackson / Applied Catalysis A: General 462–463 (2013) 121–128
Table 1
Physical properties of Rh/silica catalysts.
Catalyst
code
Pore size
(nm)
Particle size
(␮m)
Total pore
Surface area
Average metal
crystallite size
(nm)
Rhodium
dispersion
(%)
Rhodium
Rhodium
loading wt%
Pore volume
length (m g−1
)
(m2 g−1
)
surface area
(ml g−1
)
(m2 g−1
)
M01081
M01272
M01079
M01074
M01035
M01038
2.4
6.4
11.1
13.2
13.2
13.2
10.7
10
9.5
24
96.3
49.9
80,263
24,308
9814
7740
7740
594
488
344
321
321
321
1.6
1.8
1.2
2.6
3.5
2.7
71
60
94
43
32
41
7.8
6.6
10.3
4.7
3.5
4.5
2.45
2.30
2.29
2.50
2.20
2.20
0.35
0.78
0.96
1.06
1.06
1.06
7740
2. Experimental
reaction pressure. Once the vessel was pressurised the reaction
was started by switching the stirrer on at 1000 rpm: this was
taken as the zero time of the reaction. The progress of the reaction
was followed by withdrawing samples of 2.5 ml at different time
intervals along the reaction period of 6 h. The moles of hydrogen
consumed during the reaction were also monitored and recorded.
The liquid samples were analysed using a Thermo Finnigan Focus
GC equipped with an AS 3000 autosampler. The column was HP-
1701, 30 m × 0.25 mm × 1 ␮m film thickness. FID temperature was
523 K, injector temperature was 503 K, and the column temper-
ature was fixed at 433 K. The hydrogenation experiments were
performed at 3-NAP initial concentrations of 0.0075, 0.015, and
0.0225 mol l−1, temperatures of 313, 323, 333, and 343 K, hydro-
gen pressures of 2, 3, 4, and 5 barg, and catalyst weights of 0.05
and 0.1 g. The conversion of 3-nitroacetophenone and the yields
of 3-aminoacetophenone (3-AAP), 1-(3-aminophenyl) ethanol (3-
APE), and 1-(3-aminocyclohexyl) ethanol (3-ACHE) were calculated
using the following correlations:
2.1. Catalysts
The catalysts used in the hydrogenation of 3-nitroacetophenone
were rhodium catalysts supported on powder silica particles. Pow-
der silica supports were supplied and characterised by Davison
Catalysts. The active catalysts (Rh metal), which were supplied
and characterised by Johnson Matthey, were prepared using an
incipient-wetness method using aqueous rhodium chloride salts.
Six catalysts, different in pore size, particle size, surface area, and
metal crystallite size, were used to perform the hydrogenation reac-
tion under different experimental conditions. Table 1 shows the
physical properties of the as-supplied rhodium/silica catalysts [8].
2.2. Experimental setup
The hydrogenation reactions were performed using a 500 cm3
capacity Buchi autoclave stirred tank reactor supplied with an oil
heating jacket with a maximum operating temperature limit of
473 K. The temperature was measured in the liquid slurry with
accuracy of 1 K. The autoclave was designed to operate at a max-
imum operating pressure of 6 barg. The reactor was equipped with
a variable speed stirrer connected to a magnetic drive. It was pro-
vided with gas inlet/outlet tubing for inert and hydrogen gas supply
and gas venting. The reactor was connected to a Buchi press-flow
gas controller to control the pressure of the reactor. The gas con-
troller was used to measure the flow of hydrogen or inert gas to
the reactor and also to measure the hydrogen gas consumed in the
reaction. Fig. 1 shows the experimental reactor setup.
[3 − NAP]o − [3 − NAP]t
X3−NAP
Y3−AAP
Y3−APE
=
=
=
× 100
(1)
(2)
(3)
[3 − NAP]o
[3 − AAP]t
× 100
× 100
[3 − NAP]o
[3APE]t
[3 − NAP]o
2.3. Experimental procedure
In a typical hydrogenation experiment, a known quantity of
catalyst was charged with 300 ml of isopropyl alcohol (solvent)
to the reactor and heated to the desired reduction temperature
under slow stirring (300 rpm). An in situ reduction of the cata-
lyst was performed by bubbling hydrogen gas at a flow rate of
280 cm3 min−1 through this mixture for 30 minutes while stirring
at 800 rpm. A measured quantity of 3-nitroacetophenone (Aldrich,
99%) was added to 25 ml of isopropyl alcohol in a conical flask
and heated until all the 3-NAP was dissolved. After the reduc-
tion process was completed, the stirrer was turned off and the
reactor was purged with nitrogen twice and pressurised to 1 barg
pressure. The contents were then heated to the desired reac-
tion temperature under slow stirring. After reaching the desired
temperature, the stirrer was turned off and the 3-NAP solution
was added to the reactor vessel through the reactor inlet, which
was then flushed with 25 ml of isopropyl alcohol to give a total
reaction volume of 350 ml. The solution was stirred at 800 rpm
for 5 s to allow mixing. The stirrer was then turned off and the
reactor pressurised with N2 to 1 barg. A sample of 2.5 ml was
withdrawn. The reactor was de-pressurised, before being purged
twice with hydrogen then pressurised with hydrogen to the desired
Fig. 1. Schematic diagram of the experimental setup.

M.I. Abdul-Wahab, S.D. Jackson / Applied Catalysis A: General 462–463 (2013) 121–128
123
Fig. 4. Yields of reaction products with time for different catalyst weights. reaction
conditions: catalyst, M01079; temperature = 333 K; H2 pressure = 4 barg.
Fig. 2. Typical concentration–time profiles for hydrogenation of 3-NAP using
Rh/SiO2 (M01079) catalyst. Reaction conditions: catalyst weight = 0.05 g; temper-
ature = 333 K; H2 pressure = 4 barg.
The concentration of the 3-aminoacetophenone rose to a
maximum before it fell when hydrogenated further to 1-(3-
aminophenyl) ethanol, which in turn was hydrogenated via ring
hydrogenation to 1-(3-aminocyclohexyl) ethanol. The GC analysis
also detected trace levels of 3-ethyl aniline (<1%). These reac-
tion species accounted for about 90–95% of the material balance.
The hydrogenation of 3-AAP to subsequent products was limited
until the conversion of 3-NAP was nearly completed. Similarly, the
conversion of 3-APE to 3-ACHE took place only after 3-AAP concen-
trations became very small. Based on the products identified by GC
analysis, the reaction scheme is shown in Fig. 3.
The yields of 3-AAP, 3-APE, and 3-ACHE varied with the varia-
tion of the reaction conditions and catalyst loading. Fig. 4 shows
the variation of the yields with the catalyst loading at certain reac-
tion conditions. It can be noticed that doubling the catalyst weight
from 50 mg to 100 mg nearly doubles the rate of reaction for the
same reaction conditions, indicating the absence of mass trans-
fer effects in this catalyst. The yields of 3-AAP are shown in Fig. 5
for all the catalysts studied. All the catalysts have a nearly iden-
tical percentage of active metal, however as can be seen, the rate
of formation and disappearance of 3-AAP differs on each catalyst.
Generally, the rates were higher over catalysts M01038, M01035,
and M01074, which have the largest pore sizes (13.2 nm), com-
pared to the other catalysts while the slowest rate was observed
with catalyst M01081, which has the smallest pore size (2.4 nm).
Fig. 6 shows the yield of 3-APE for all the catalysts studied. Again
the rates of formation and consumption of 3-APE were higher in
[3ACHE]t
[3 − NAP]o
Y3−ACHE
=
× 100
(4)
where, X is the conversion, Y is the yield, [3-NAP]o is the initial con-
centration of 3-NAP, [3-NAP]t, [3-AAP]t, [3-APE]t, and [3-ACHE]t,
are the concentrations of 3-NAP, 3-AAP, 3-APE, and 3-ACHE respec-
tively at any time.
3. Results
In the present study, hydrogenation of 3-nitroacetophenone
took place through a series of consecutive reactions. Preliminary
experiments were performed to study the product distribution
and selectivity behaviour using the Rh/SiO2 catalysts. The effect of
catalyst particle size, metal crystallite size, and pore size was inves-
tigated for a range of Rh/SiO2 catalysts at different temperature,
hydrogen pressure and catalyst loading conditions.
3.1. Reaction products distribution
Both the choice of catalyst and the reaction parameters have
significant effects on the yields of the hydrogenation products. A
typical concentration–time profile of various components involved
in the hydrogenation of 3-NAP using Rh/SiO2 is shown in Fig. 2. It
was observed that 3-NAP underwent hydrogenation of the NO2
group to form 3-aminoacetophenone.
Fig. 3. Reaction scheme for hydrogenation of 3-NAP over Rh/SiO2.

124
M.I. Abdul-Wahab, S.D. Jackson / Applied Catalysis A: General 462–463 (2013) 121–128
Fig. 5. Yields of 3-aminoacetophenone for different catalysts. Reaction conditions:
catalyst weight = 0.05 g; temperature = 333 K, H2 pressure = 4 barg.
Fig. 8. Concentration–time profiles for hydrogenation of 3-NAP using different 3-
NAP initial concentrations. Reaction conditions: catalyst weight (M01079) = 0.05 g;
temperature = 333 K; H2 pressure = 3 barg.
has an effect on the reaction rate, four experiments were carried
out with different stirrer speeds (600, 800, 1000 and 1200 rpm)
while all other reaction conditions were held constant. Hydrogen
consumption was almost identical in all tests indicating that gas-
film mass transfer had no significant effect on the rate of reaction.
Liquid–solid mass transfer limitations were tested by doubling the
catalyst weight for the same reaction conditions. The results show
that doubling the catalyst amount approximately doubles the rate
of reaction, which indicates the absence of liquid–solid mass trans-
fer effects. Pore diffusion, in general, is not likely to take place in
liquid phase reactions because pore diffusion (mainly Knudsen dif-
fusion) normally relates to gases rather than liquids since the mean
free path for molecules of liquid is very small and may be close to the
diameter of the molecule itself [9]. In the present study, the lowest
pore size was in catalyst M01081 (2.4 nm), which is approximately
8 times the molecular diameter (0.309 nm) of 3-NAP. However, this
catalyst showed less activity for the hydrogenation reaction when
compared to the other catalysts, which indicate the possibility of a
mass transfer contribution in the kinetics of the reaction.
Fig. 6. Yields of 1-(3-aminophenyl) ethanol for different catalysts. Reaction condi-
tions: catalyst weight = 0.05 g; temperature = 333 K, H2 pressure = 4 barg.
the case of catalysts M01038, M01035, and M01074, compared to
other catalysts. These catalysts also showed a larger tendency for
ring hydrogenation to 3-ACHE. Catalyst M01081, of course, is the
slowest in producing 3-aminophenyl ethanol and also in the ring
hydrogenation. The same trend was noticed for the production of
1-(3-aminocyclohexyl) ethanol (Fig. 7).
3.3. Effect of 3-NAP initial concentration
In order to investigate the effect of the initial concentration
of 3-NAP on the rate of reaction, the hydrogenation was carried
out using 3-NAP with initial concentrations of 0.0075, 0.015, and
0.0225 mol l−1. Fig. 8 shows the concentration–time profiles of
3-NAP using catalyst M01079, with all other reaction conditions
kept constant. One must be careful in calculating the reaction rate
when using different concentrations, as the instantaneous rates will
differ as the reaction proceeds. Therefore, the initial rate over a
short period of time must be calculated. The hydrogen consump-
tion profile indicates that H2 uptake was nearly the same for the
first 15 minutes of the reaction times. The rates were approxi-
3.2. Effects of mass transfer
In order to examine the effect of mass transfer on the hydro-
genation reaction, a series of tests were performed to ensure that
there was no mass transfer resistance interference with the intrin-
sic rate measurements. To investigate whether film mass transfer
mately identical for the three initial concentrations at 1.08 × 10−4
,
1.15 × 10−4, and 1.17 × 10−4 mol l−1min−1, indicating that the ini-
tial concentration of 3-NAP has a negligible effect on the rate of
reaction and the hydrogenation reaction can be considered as zero
order with respect to 3-NAP concentration.
3.4. Effect of temperature and hydrogen pressure
Hydrogenation of 3-NAP was carried out at different tempera-
tures, 313, 323, 333, and 343 K and different hydrogen pressures,
2, 3, 4, and 5 barg. A significant enhancement in the activity was
observed with the increase in the reaction temperature (Fig. 9).
The complete conversion of 3-nitroacetophenone was reached after
300 min at 313 K, while it was reached after 60 min at 343 K for the
Fig. 7. Yields of 1-(3-aminocyclohexyl) ethanol for different catalysts. Reaction con-
ditions: catalyst weight = 0.05 g; temperature = 333 K, H2 pressure = 4 barg.

M.I. Abdul-Wahab, S.D. Jackson / Applied Catalysis A: General 462–463 (2013) 121–128
125
Fig. 9. Concentration profiles of 3-NAP at different temperatures using catalyst
M01079. Reaction conditions: catalyst weight = 0.05 g; H2 pressure = 4 barg.
Fig. 12. Yields of 3-ACHE at different temperatures using catalyst M01079. Reaction
conditions: catalyst weight = 0.05 g; H2 pressure = 4 barg.
Fig. 13. Concentration of 3-NAP for different hydrogen pressure using catalyst
M01079. Reaction conditions: catalyst weight = 0.05 g; temperature = 333 K.
Fig. 10. Yields of 3-AAP at different temperatures using catalyst M01079. Reaction
conditions: catalyst weight = 0.05 g; H2 pressure = 4 barg.
were used to estimate the apparent activation energies for all the
catalysts used. Table 2 shows that rhodium/silica catalysts with dif-
ferent physical properties have different activation energies for the
hydrogenation reaction.
The effect of hydrogen pressure was tested by carrying out
experiments at different hydrogen pressures (2, 3, 4, and 5 barg)
to examine the effect of the change in pressure on the rate of the
hydrogenation reaction. Fig. 13 shows that the hydrogen pressure
influenced the rate of 3-NAP hydrogenation. Table 3 shows the
order of reaction with respect to the change of hydrogen pressure
for all the catalyst used. The orders lies in the range of 0.4–0.5 for
catalyst M01079. The effect of temperature on the product yields is
shown in Figs. 10–12. The yield of 3-AAP decreased as the reaction
temperature increased from 313 to 343 K due to the formation of
the alcohol. The behaviour of 3-APE was similar as its concentration
rose and then fell as 3-ACHE was formed. The drop in concentra-
tion was more rapid with the increase in reaction temperature. The
ring hydrogenation of 1-(3-aminophenol) ethanol was relatively
slow at low temperatures but became faster at higher temper-
atures indicating that if it is desired to have high selectivity of
3-APE, the reaction should be carried out at low temperatures (less
than 323 K). Arrhenius plots of the hydrogenation of 3-NAP to 3-
AAP with varying reaction temperatures (313, 323, 333, and 343 K)
Table 2
Activation energies for the hydrogenation of 3-NAP.
Catalyst
Activation energy (kJ mol−1
)
M01081
M01272
M01079
M01074
M01038
M01035
68.4
47.3
43.9
40.0
57.6
43.4
Table 3
Reaction order with respect to hydrogen pressure.
Catalyst
Reaction order
M01081
M01272
M01079
M01074
M01038
M01035
1.3
0.4
0.4
0.4
0.5
0.4
Fig. 11. Yields of 3-APE at different temperatures using catalyst M01079. Reaction
conditions: catalyst weight = 0.05 g; H2 pressure = 4 barg.

126
M.I. Abdul-Wahab, S.D. Jackson / Applied Catalysis A: General 462–463 (2013) 121–128
Fig. 14. Hydrogen consumptions for catalysts with different particle sizes. Reaction
conditions: catalyst loading = 0.05 g; temperature = 333 K; H2 pressure = 4 barg.
Fig. 16. Turnover frequency versus metal crystallite size at different reaction tem-
peratures. Reaction conditions: catalyst weight = 0.05 g; H2 pressure = 4 barg.
3.6. Effect of catalyst metal crystallite size
Metal crystallite size is normally used as a measure of metal
dispersion on the silica support. Metal dispersion was calculated as
the ratio of the moles of surface metal to the total moles of metal
(Eq. (5)).
SARhMwt
A
Dispersion% =
× 100
RhNWRh
(5)
where SARh is the metal surface area per unit catalyst weight
(m2 g−1), ARh is the area of Rh atom (m2) which is taken as
7.52 × 10−2 m2 [8], N is Avogadro number (6.02 × 1023), and WRh
is the percentage of the catalyst metal loading. Table 1 shows the
percentage dispersion of the various rhodium catalysts. The hydro-
genation activity of the catalysts was compared using the turnover
frequency rather than the reaction rate because the catalysts had
different metal surface areas. TOF is defined as the number of
molecules of hydrogen consumed per surface Rh atom per minute.
It is calculated by using the following equation:
Fig. 15. Concentration–time profiles for hydrogenation of 3-NAP using Rh/SiO2,
for catalysts M01074, M01038, and M01035. Reaction conditions: catalyst
weight = 0.05 g; temperature = 333 K; H2 pressure = 4 barg.
all the catalysts except for catalyst M01081, which has the lowest
pore size (2.4 nm) and had an order of 1.3.
nH2 reacted
TOF =
(6)
nRh
where nH2
is the number of moles of hydrogen reacted per
reacted
3.5. Effect of catalyst particle size
minute per gram of catalyst and nRh is the number of moles of sur-
face rhodium per gram of catalyst. Catalysts, M01074, M01038, and
M01035, which have identical surface area and pore diameters but
differ in metal crystallite size and catalyst M01079 which differs
slightly in pore size (Table 1), were used to investigate the effect
of metal crystallite size (or metal dispersion) on the activity of the
catalysts. A plot of TOF vs. metal crystallite size at different reaction
temperatures is shown in Fig. 16. There is an obvious increase in
TOF with the metal crystallite size of the catalysts which becomes
steeper with increasing temperature.
Three of the Rh/SiO2 catalysts; M01074, M01038, and M01035
were tested for the effect of particle size on the rate of hydrogena-
tion reaction. These catalysts have identical pore sizes (13.2 nm)
and surface area (321 m2 g−1) but different particle sizes (24, 49.9,
and 96.6 ␮m respectively). Fig. 14 shows the hydrogen consump-
tion curve and Table 4 shows the initial rate of the reaction. It can
be observed that the difference in particle sizes of the catalysts
had only a minor effect on the initial rate of the hydrogenation
reaction. Although the reaction profiles were similar for all the
reactions performed (Fig. 15), there was a slight increase in the ini-
tial rate when going from catalyst M01074 (particle size = 24 ␮m)
to M01038 (particle size = 96.6 ␮m). The yield of 3-AAP with con-
version shows a minor effect of particle size on the yield even for
different temperatures.
3.7. Effect of catalyst pore size
Catalysts M01081, M01272, and M01079 were used to investi-
gate the effect of the pore size on the activity of the catalysts in
the hydrogenation reaction. To decouple the effect of differences in
average metal crystallite sizes of the above catalysts, the method
applied by Hindle et al. [8] in which the metal crystallite size was
normalised to 1 nm was adopted. Fig. 17 shows a plot of TOF values
versus pore size at different reaction temperatures. As expected,
catalyst M01081, which had the smallest pore size (2.4 nm) gave
the smallest TOF value compared to the other catalysts, which gave
nearly identical values. It can be concluded that the pore size has a
significant effect on the catalyst M01081 and only minor effects on
the other two catalysts. The effect of temperature on the catalyst
Table 4
Initial reaction rate for Rh/SiO2 catalysts of different particle sizes.
Catalyst
Cat
Cat
Cat
M01074
PS = 24 ␮m
M01038
PS = 49.9 ␮m
M01035
PS = 96.6 ␮m
Initial rate
5.25
5.62
5.52
(mmol min−1 g−1
)

M.I. Abdul-Wahab, S.D. Jackson / Applied Catalysis A: General 462–463 (2013) 121–128
127
that the formation of 3-AAP, 3-APE, and 3-ACHE is consecutive
reactions. No useful literature was found concerning the forma-
tion of 1-(3-aminocyclohexyl) ethanol by ring hydrogenation of
1-(3-aminophenyl) ethanol. When the concentration of 3-APE
became large enough, especially at high temperatures, a small
amount of 3-ethylaniline (<1%) was produced. This is could be
due to the dehydration reaction of 1-(3-aminophenyl) ethanol to
give 3-aminostyrene, which would then be rapidly transformed to
3-ethylaniline by hydrogenation. However no 3-aminostyrene was
observed and the reaction may be direct reduction of the alcohol.
The rate of the hydrogenation reactions was affected by the
physical characteristics of catalysts and the operating conditions
such as temperature and pressure. Results show that increasing
the initial concentration of 3-NAP from 0.0075 to 0.0225 mol l−1
did not alter the initial rate of the reaction. Therefore, the hydro-
genation reaction order with respect to 3-NAP concentration is zero
order. Hydrogen pressure has also a significant effect on the rate of
the reaction and the yields. The initial rate of 3-NAP hydrogena-
tion increased with an increase of the initial hydrogen pressure
in the reactor. This result agrees with many previous studies on
the hydrogenation of aromatic nitro species [1,10–12]. With most
of the catalysts the reaction order in hydrogen is ∼0.5 (Table 3).
Such a value is consistent with a Langmuir–Hinshelwood descrip-
tion of the reaction kinetics with hydrogen being dissociated and
weakly adsorbed relative to the 3-NAP. The single catalyst, which
gives a different value for the reaction order in hydrogen, is M01081
with a value that approximates to first order. This catalyst has the
smallest pore size and, as will discussed later, shows mass transfer
control effects in the hydrogenation. Typically mass transfer con-
trolled reactions exhibit first order kinetics; hence the value of ∼1
for the reaction order with respect to hydrogen pressure is not sur-
prising. Therefore for this catalyst, only kinetic terms modified by
mass transfer control will be observed. Changing the temperature
of the reaction allowed a determination of the activation energy for
the hydrogenation of the 3-NAP to 3-AAP (Table 2). The value that
is significantly different from the rest is that of catalyst M01081.
It would be expected that the effect of mass transfer control on
the reaction would lower the observed activation energy to a value
≤20 kJ mol−1, however the reverse is found with an observed acti-
vation energy of ∼68 kJ mol−1. This behaviour may be related to the
exothermic nature of the reaction, which can result in a tempera-
ture difference between the bulk and the surface of the catalyst
particle leading to an enhanced value for the apparent activation
energy [17].
The investigation of the effect of average metal crystallite size
or metal dispersion on the turnover frequency, which showed that
the turnover frequency decreased with an increase in rhodium dis-
persion, is consistent with previously reported observations, for
example, platinum supported catalysts showed that the turnover
frequency increased with a decrease in platinum dispersion along
with an increase in the acidity of support materials [18]. Hydro-
genation of para-toluidine on Rh/SiO2 catalysts also displayed a
decrease in catalyst activity with the increase in metal disper-
sion [19]. The antipathetic relationship between metal dispersion
and TOF can be attributed to the nature of the rhodium metal
crystals and the specificity of the reaction site. The structure of
rhodium crystals is of face-centred cubic (FCC), which is a close-
pack structure with atomic packing factor (APF) of 0.74. When
the metal crystal size increases, the number of face surface atoms
increases at the expense of the edge and corner sites of the struc-
ture. The increase of turnover frequency with the increase of the
metal particle size suggests that the hydrogenation reaction takes
place on the plane face surface as opposed to edge and corner
sites. Matsuhashi et al. [20] in their comprehensive study on the
effect of metal dispersion in many catalytic reactions over plat-
inum catalysts concluded that the relation between the TOF and
Fig. 17. Turnover frequency versus catalyst pore size at different reaction temper-
atures. Reaction conditions: catalyst weight = 0.05 g; H2 pressure = 4 barg.
activity is clear as TOF increases with increasing of reaction tem-
perature; however, the rate of increase of TOF becomes larger at
the higher temperatures.
4. Discussion
The aim of the present work was to study the hydrogena-
tion of 3-nitroacetophenone and more particularly, the production
of 3-aminoacetophenone, 1-(3-aminophenyl) ethanol and 1-(3-
aminocyclohexyl) ethanol by the catalytic hydrogenation of
3-nitroacetophenone using different rhodium/silica catalysts in a
semi-batch reactor at different reaction conditions. 3-NAP hydro-
genation is a complex multi-step reaction due to competitive
hydrogenation between nitro, carbonyl, and phenyl groups in one
molecule. Prior to discussing the results, it is important to note
that there is no significant information about the mechanism of
hydrogenation of 3-nitroacetophenone in the literature. Therefore,
some analogous reactions are examined to reveal useful informa-
tion regarding the formation of products.
In the previous studies of hydrogenation of 3-
nitroacetophenone [1–3,5] and other studies on the hydrogenation
of compounds containing nitro group such as nitrobenzene
[1,10–13], hydrogenation of the nitro group is expected to occur
faster than other functional groups, due to the strong electroneg-
ativity of the nitro group, which comes from the combined action
of the two electron-deficient oxygen atoms bonded to the par-
tially positive nitrogen atom; hence 3-aminoacetophenone was
expected, and found, as the first main product. The hydrogenation
of 3-AAP is more complex due to the competitive hydrogena-
tion between phenyl and carbonyl groups. Hydrogenation of
acetophenone [14], in the absence of the amine functionality,
showed different possible products depending on the catalysts
used, for example products detected over a Pd-based catalyst [15]
were 1-phenylethanol and ethylbenzene, while over a Pt-based
catalyst 1-phenylethanol, ethylbenzene, methylcyclohexylketone,
cyclohexylethanol and ethylcyclohexane were all observed [16].
Examination of the present experimental results shows that the
hydrogenation of the carbonyl group in 3-AAP has priority over
hydrogenation of the phenyl group over silica supported Rh cata-
lysts. The ring hydrogenation of 3-AAP did not take place during
the reaction, since there was no 1-methyl-(3-aminocyclohexyl)
ketone detected in the products. Therefore, Rh catalysts can be
considered as good catalysts for the selective hydrogenation of
the carbonyl group. The concentration of 3-aminoacetophenone
increases to a maximum value and then decreases with reaction
time, as the concentration of 1-(3-aminophenyl) ethanol increases
and then decreases, while finally the concentration of 1-(3-
aminocyclohexyl) ethanol slowly grows with time. This means

128
M.I. Abdul-Wahab, S.D. Jackson / Applied Catalysis A: General 462–463 (2013) 121–128
platinum dispersion was negative in a number of reactions such as,
combustion of propane, oxidation of CO, hydrogenations of ethyl-
ene and naphthalene, hydrodechlorination of trichloroethane and
hydrodesulfurization of thiophene. They indicated that the activity
increases with decreasing metal particle size when the active sites
exist on the edges or corners of metal surfaces, which increases with
increasing dispersion. Our interpretation of the results from the
present study is in keeping with this conclusion. Nevertheless care
must be taken as this explanation has in-built assumptions such
as the need for the metal crystallites to have complete outer layer.
Many crystallites will have defects and will not have complete crys-
tal structures. Another property that changes with metal crystallite
size is the electronic structure of the metal. In a recent publication
[21] it has been proposed that below 10 nm it is the electronic fac-
tor that will have the largest effect on a surface sensitive reaction
rather than the geometric factor and in our study all the catalysts
have average metal crystallite sizes well below 10 nm. Indeed in
small metal crystallites geometric and electronic factors are inex-
tricably linked and both will play a role in structure sensitivity of
small metal crystallites [21–24].
Although the liquid–solid reaction systems can usually be con-
sidered safe from pore diffusion effects, the results show that
catalyst M01081with the smallest pore size (2.4 nm) has the lowest
activity among the catalysts studied, which may indicate the exist-
ence of some diffusion resistance within this catalyst. The result of
the variation of TOF with pore size, shown in Fig. 17, confirms this
interpretation. Other catalysts which have pores sizes of 6.4, 11.1,
and 13.2 nm show nearly identical TOF values. Therefore, it can be
concluded that the pore diffusion can be avoided in the hydrogena-
tion of 3-NAP, when using Rh/SiO2 catalysts with pore sizes larger
than 6.4 nm.
a marked increase at high temperatures (more than 333 K). The
results show that the physical properties of the catalyst support
have significant effect on the activity of the catalyst. The effects of
dispersion, particle size and pore size of catalysts have been pre-
sented. It is observed that with the catalyst particle size range used
in the present study (9.5–96.3 ␮m), the size of catalyst particles
has only minor effect on the activity. The pore size of the cata-
lyst support however has a marked effect on TOF for the catalyst
of pore size less than 6.4 nm while the catalysts with pore sizes
larger than 6.4 nm show nearly identical TOF values. The metal dis-
persion shows an antipathetic relationship with the activity of the
catalyst, such that catalysts with higher dispersion values or smaller
metal crystallite sizes show lower TOF values. This behaviour can be
related to changes in the electronic and structural make up of small
metal crystallites and may indicate that the reaction takes place on
the faces of the FCC rhodium close-pack crystalline structures.
Acknowledgments
Financial support from the Institute of International Education
(IIE) and SRF Iraq-Project are gratefully acknowledged.
References
[1] S.D. Jackson, R. McEwan, R.R. Spence, in: M.J. Prunier (Ed.), Catalysis of Organic
Reactions, 22nd Conference, CRC Press, Boca Raton, 2008, pp. 79–86.
[2] W. Michalowicz, L. Haven, U.S Patent 4021487, 1977.
[3] J. Lomartire, N.J. Summit, US Patent 2,680,136, 1954.
[4] W. Emerson, R.A. Helmsch, U.S Patent 2,683,745, 1954.
[5] S.W. Tinsley, S. Charleston, U.S. Patent 2,797,244, 1957.
[6] P.N. Rylander, U.S. Patent 3,423,462, 1969.
[7] J.M. Hawkins, T.W. Makowski, Org. Process Res. Dev. 5 (2001) 328–330.
[8] K.T. Hindle, S.D. Jackson, D. Stirling, G. Webb, J. Catal. 241 (2006) 417–425.
[9] O. Levenspiel, Chemical Reaction Engineering, 3rd ed., Wiley, New York, 1999,
pp. 378.
[10] E.A. Gelder, University of Glasgow, 2005, PhD Thesis.
[11] E.A. Gelder, S.D. Jackson, C.M. Lok, Catal. Lett. 84 (3/4) (2002) 205–208.
[12] C. Li, Y. Chen, W. Wang, Appl. Catal. A 119 (1994) 185–194.
[13] P. Sangeetha, K. Shanthi, K.S. Rama Rao, B. Viswanathan, P. Selvam, Appl. Catal.
A 353 (2009) 160–165.
[14] X.B. Zhang, React. Kinet. Mech. Catal. 102 (2011) 417–421.
[15] C. Chen, H. Chen, Appl. Catal. A 260 (2004) 207–213.
[16] C. Chen, H. Chen, W. Cheng, Appl. Catal. A 248 (2003) 117–128.
[17] M.E. Davis, R.J. Davis, Fundamentals of Chemical Reaction Engineering,
McGraw-Hill Co., New York, 2003, pp. 184.
[18] Y. Yazawa, H. Yoshida, T. Hattori, Appl. Catal. A 237 (2002) 139–148.
[19] K.T. Hindle, S.D. Jackson, G. Webb, in: J.R. Sowa Jr. (Ed.), Catalysis of Organic
Reactions, 20th Conference, CRC Press, Boca Raton, 2005, pp. 77–84.
[20] H. Matsuhashi, S. Nishiyama, H. Miura, K. Eguchi, K. Hasegawa, Y. Iizuka, A.
Igarashi, N. Katada, J. Kobayashi, T. Kubota, T. Mori, K. Nakai, N. Okazakim, M.
Sugioka, T. Umekio, Y. Yazawa, D. Lu, Appl. Catal. A 272 (2004) 329–338.
[21] P.E. Strizhak, A.I. Trypolskyi, G.R. Kosmambetova, O.Z. Didenko, T.N. Gurnyk,
Kinet. Catal. 52 (2011) 128–138.
5. Conclusion
Silica-supported rhodium catalysts were used in the investiga-
tion of the effects of catalyst physical properties and reaction con-
ditions on the hydrogenation of 3-nitroacetophenone. The results
show that the reaction is consecutive with 3-aminoacetophenone,
1-(3aminophenyl) ethanol, and 1-(3-aminocyclohexyl) ethanol the
major products. The reactivity of the functional groups to hydro-
genation over Rh/SiO2 catalysts is nitro > carbonyl > phenyl. The
initial hydrogenation reaction is zero order with respect to 3-
nitroacetophenone concentrations and half order (0.5) with respect
to hydrogen. The rate of reaction increases with increasing temper-
ature resulting in activation energies between 40 and 60 kJ mol−1
.
The rate of formation and consumption of intermediates varies with
temperature and hydrogen pressure. A full kinetic analysis of the
sequential reactions will be published in a subsequent paper. The
yield of the final product, 1-(3-aminocyclohexyl) ethanol, shows
[22] G.C. Bond, Catal. Rev. 50 (2008) 532–567.
[23] G.C. Bond, Chem. Soc. Rev. 20 (1991) 441–475.
[24] G.C. Bond, Surf. Sci. 156 (1985) 966–981.